Methadone maintenance is an effective tool for reducing HIV risk and incidence among IDUs. All too often, however, individuals in methadone treatment continue to use opiates and other drugs, usually with a close friend or family member not currently enrolled in treatment. One response to this problem is to encourage at least one member of the methadone client's personal drug-using network to also enter treatment, thereby reducing the number of individuals presenting drug use opportunities. In this pilot study, 268 new methadone treatment entrants (Index Participants) will be interviewed and invited to bring in an opiate using sexual partner, family member, or close friend (Partner participants) described as belonging to their current personal injection drug-use network for interview. We anticipate that half of the 268 will be able to recruit a Partner for interview. The resulting 134 Partner participants will be randomly assigned to either the Intervention condition (offered 13 weeks of free methadone maintenance) or to a Control condition (no offer of free treatment). Self-report, urinalysis records in the clinical charts, and other biological indicator data (hair samples) will be collected at baseline and 4 months post-randomization to assess treatment outcomes, change in HIV risk behaviors, and change in personal drug-using network characteristics for both Index participants and their designated Partner participants. This two-group, random assignment pilot study is designed to address the following specific aims: Specific Aim 1: to determine if offering free drug treatment to one Partner will improve treatment outcomes and reduce HIV risk behaviors among Index participants. Specific Aim 2: to determine if offering free drug treatment to the Partner will result in reductions in Partner participant's HIV risk behaviors and drug use. Specific Aim 3: to determine if offering free drug treatment to one Partner will significantly alter the size, density, and stability of the personal drug-using network of the Partner, and ultimately of the referring Index. The proposed study addresses a practical public health problem, that ongoing drug use during methadone treatment tends to be initiated by close friends, family members, or significant others. The overall goal of the proposed study is to demonstrate that treatment outcomes can be improved and HIV risk behaviors reduced by involving at least two network members in drug treatment simultaneously.